HAL Id: hal-02606317
https://hal.inrae.fr/hal-02606317
Submitted on 16 May 2020
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Water Framework Directive Intercalibration:
Central-Baltic Lake Fish fauna ecological assessment methods. Part A: Descriptions of fish-based lake
assessment methods
D. Ritterbusch, Christine Argillier, J. Arle, W. Bialokoz, J. Birzaks, P.
Blabolil, J. Breine, H. Draszkiewicz-Mioduszewska, N. Jaarsma, Y. Karottki, et al.
To cite this version:
D. Ritterbusch, Christine Argillier, J. Arle, W. Bialokoz, J. Birzaks, et al.. Water Framework Directive
Intercalibration: Central-Baltic Lake Fish fauna ecological assessment methods. Part A: Descriptions
of fish-based lake assessment methods. irstea. 2017, pp.95. �hal-02606317�
David Ritterbusch, Christine Argillier, Jens Arle, Witold Białokoz,
Janis Birzaks, Petr Blabolil, Jan Breine, Hanna Draszkiewicz-Mioduszewska, Nico Jaarsma, Ivan Karottki, Teet Krause, Jan Kubečka, Torben Lauridsen, Maxime Logez, Anthony Maire, Anu Palm,
Graeme Peirson, Milan Říha, Jacek Szlakowski, Tomas Virbickas, Sandra Poikane
Part A: Descriptions of fish-based lake assessment methods
Water Framework Directive Intercalibration:
Central-Baltic Lake Fish fauna ecological assessment methods
2017
EUR 28022 EN
Water Framework Directive Intercalibration:
Central-Baltic Lake Fish fauna ecological assessment methods
Part 1: Descriptions of fish-based lake assessment methods
1
This publication is a Technical report by the Joint Research Centre (JRC), the European Commission’s science and knowledge service. It aims to provide evidence-based scientific support to the European policymaking process. The scientific output expressed does not imply a policy position of the European Commission. Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use that might be made of this publication.
Contact information Name: Sandra Poikane
Address: Via E. Fermi 2749, Ispra (VA), 21027 Italy Email: [email protected]
Tel.: +39 0332 789720
JRC Science Hub https://ec.europa.eu/jrc
JRC101684
EUR 28022 EN
PDF Print
ISBN 978-92-79-59935-4 ISSN 1831-9424 doi:10.2791/084375 ISBN 978-92-79-59936-1 ISSN 1018-5593 doi:10.2791/396601
Luxembourg: Publications Office of the European Union, 2017
© European Union, 2017
The reuse of the document is authorised, provided the source is acknowledged and the original meaning or message of the texts are not distorted.
The European Commission shall not be held liable for any consequences stemming from the reuse.
How to cite this report: D. Ritterbusch, C. Argillier, J. Arle, W. Białokoz, J. Birzaks, P. Blabolil, J. Breine, H. Draszkiewicz-Mioduszewska, N.
Jaarsma, I. Karottki, T. Krause, J. Kubečka, T. Lauridsen, M. Logez, A. Maire, A. Palm, G. Peirson, M. Říha, J. Szlakowski, T. Virbickas, S. Poikane.
Water Framework Directive Intercalibration: Central-Baltic Lake Fish fauna ecological assessment methods; Part A: Descriptions of fish-based lake assessment methods; EUR 28022 EN; doi:10.2791/396601.
All images © European Union 2017
2
Table of contents
Acknowledgements ... 3
Abstract ... 4
A1 Introduction ... 5
A2 Overview ... 6
A2.1 Status of LFI assessment systems ... 6
A2.2 Fish sampling methods ... 6
A2.3 Typology... 7
A2.4 Metrics ... 8
A3 WFD compliance ... 10
A4 Intercalibration feasibility... 12
A4.1 Typology as a restricting factor ... 12
A4.2 Pressure criteria as restricting factors ... 13
A4.3 Sampling comparability ... 13
A4.4 Metric comparability ... 14
A4.5 Definition of reference conditions and class boundary setting ... 14
A5 Summary and conclusion... 16
References ... 17
Annex A I: National descriptions of Lake fish assessment systems ... 21
Belgium-Flanders ... 21
Czech Republic ... 28
Denmark ... 40
Estonia ... 44
France ... 49
Germany ... 59
Lithuania ... 62
Netherlands ... 67
Poland ... 74
United Kingdom ... 84
Annex A II: comments on the compliance criteria ‘age structure’ and ‘sensitive species’ ... 85
List of abbreviations and definitions ... 89
List of figures ... 91
List of tables ... 93
3
Acknowledgements
The work of the group leader (David Ritterbusch, DE) was funded by the German federal countries’ program of financing ‘Water, Soil and Waste’.
The Flemish participant Jan Breine would like to thank the Flemish Environment Agency (VMM) for financial support.
The Czech participants were supported by project CEKOPOT (CZ.1.07/2.3.00/20.0204), co- financed by the European Social Fund the state budget of the Czech Republic.
The work of Nico Jaarsma (NL) was funded by the Ministry of Infrastructure and the
Environment (I&M).
4
Abstract
The European Water Framework Directive (WFD) requires the national classifications of good ecological status to be harmonised through an intercalibration exercise. In this exercise, significant differences in status classification among Member States are harmonized by comparing and, if necessary, adjusting the good status boundaries of the national assessment methods.
Intercalibration is performed for rivers, lakes, coastal and transitional waters, focusing on selected types of water bodies (intercalibration types), anthropogenic pressures and Biological Quality Elements. Intercalibration exercises are carried out in Geographical Intercalibration Groups - larger geographical units including Member States with similar water body types - and followed the procedure described in the WFD Common Implementation Strategy Guidance document on the intercalibration process (European Commission, 2011).
The Technical report on the Water Framework Directive intercalibration describes in detail how the intercalibration exercise has been carried out for the water categories and biological quality elements. The Technical report is organized in volumes according to the water category (rivers, lakes, coastal and transitional waters), Biological Quality Element and Geographical Intercalibration group.
This volume addresses the intercalibration of the Lake Central-Baltic Fish ecological assessment methods.
Part A: This document comprises an overview and detailed descriptions of fish-based lake ecological assessment methods.
Part B describes the construction of multiple pressure index in the Central-Baltic region.
Part C describes the procedure and results of the boundary harmonisation of national fish-
based lake assessment systems.
5
A1 Introduction
In the Central-Baltic Geographical Intercalibration Group nine Member States submitted 10 lake fish-based assessment methods for the intercalibration: Belgium-Flanders, Czech Republic, Denmark, Estonia, France, Germany, Lithuania, the Netherlands, and Poland (two methods).
After evaluation of the WFD compliance and Intercalibration feasibility all methods were included in the intercalibration exercise. Intercalibration “Option 2” was used - indirect comparison of assessment methods using common pressure metrics (TAPI index).
This document provides the overview of the national fish-based lake assessment methods systems, includes the first steps of the intercalibration according to the IC Guidance:
1. Overview of the national
fish-based assessment methods
: concepts, typologies, metrics, scoring;2. The WFD compliance check - are all
lake fish-based assessment methods
in line with the WFD requirements?3. The intercalibration feasibility check - is there any chance that Intercalibration might be successful?
The detailed national descriptions can be found in Annex I.
6
A2 Overview
A2.1 Status of national fish-based lake assessment systems
Nearly all Member States (MS) had developed national methods to assess the ecological status of lakes with fish (Table A1). Short descriptions of all systems are available in the annex A1. Comprehensive descriptions exist for Czech Republic (B
OROVECet al. 2014) - in Czech, Denmark (S
ØNDERGAARDet al. 2013) - in Danish, France (A
RGILLIERet al. 2013) and the Netherlands (J
AARSMA2007).
Table A1: Status of Lake fish assessment systems (no method / under development / intercalibratable / official) and year of finalization or expected finalization.
MS System name Status Year Comment
BE-F Intercalibratable 2012
CZ CZ-FBI official, national 2013/2014
DE DeLFI Intercalibratable 2011 Finalized but not official DK Danish Lake Fish
Index
Intercalibratable 2011
EE LAFIEE Intercalibratable 2009 Under revision
FR 2IL Intercalibratable,
national
2013/2014
LT LZIE Intercalibratable 2013 Finalized but not official NL VISMAATLAT Intercalibratable,
national
2007 Minor changes made in 2012
PL LFI+
LFI EN
Intercalibratable Intercalibratable
2011 2013
Two methods: one based on fisheries statistics (LFI+), another developed for EN 14757
UK* Under development Unclear
LV No method Unclear
SK No method Not
expected
UK: England and Wales are part of the Central-Baltic (CB) Geographical Intercalibration Group (GIG). The Irish Lake Fish system and the CB methods will be checked for applicability. Gillnet fishing is not an option because of public relation issues.
A2.2 Fish sampling methods
The following Table A2 shows the methods used for fish stock assessment in the CB GIG. The
randomized multi-mesh gillnet standard (EN 14757 2005) is widespread, but not used in all
MS.
7
Table A2: Fish sampling methods in the MS. Second column shows the gear used for metric calculation in the national LFI systems.
Nr MS Gear for LFI metrics Database** Additional gear used ***
1 BE-F Fyke + electrofishing No
2 CZ ENmod* Yes (24) in
2014
Electrofishing, hydroacoustics
3 DE EN / ENmod Yes (75) Electrofishing
4 DK EN Yes (113) Electrofishing
5 EE ENmod Yes (24) Fyke nets, electrofishing
6 FR EN Yes (40)
7 LT ENmod Yes (46)
8 LV Gillnets, trammel nets, statistics 9 NL Trawling, seine, electrofishing No 10 PL - LFI+
PL - EN
Fisheries statistics (seine, gillnet, fyke)
En
No Electrofishing
11 UK Plans to use environmental
DNA
*EN is EN 14757, ENmod is comparable to EN 14757 but modified (explanations below).
**‘Database’ shows if data was submitted to the CEMAGREF cross-GIG database with the number of lakes in parentheses. ***Additional gear shows methods used for sampling, but not for metrics in the national LFI.
Comments:
- In CZ the standard 12-mesh sizes gillnets were extended by four larger mesh sizes (70, 90, 110 and 135 mm) for capturing bigger fish (Š
MEJKALet al. 2015);
- In DE some lakes were fished twice, in spring and autumn with half EN effort each.
This is not done any more and a single full EN campaign is used to obtain fish data.
Large mesh sizes are included in the netting. Electrofishing data is always present but not included in the system;
- EE uses the EN standard with additional large mesh sizes;
- LT uses the EN standard with reduced and modified mesh sizes: 14, 18, 22, 25, 30, 40, 50, 60 mm, nets are highly modified;
- LV focuses on species of commercial or recreational fishing.
A2.3 Typology
The typologies used in the CB GIG are shown in Table A3. DK and NL apply the common
intercalibration types developed in IC phase I (P
OIKANE2009). The other MS deviate from this
typology, although the distinction between shallow polymictic and deeper stratified lakes
apparently is the most important criterion. For IC purposes, the MS agreed to adopt the
German typology (1
stCB IC meeting, 2010, Berlin).
8
Table A3: National typologies used in the CB Lake Fish intercalibration.
MS National types Details and comments
BE-F Standing waters Lakes, ponds and canals
CZ Polymictic Stratified
For HMWB and AWB (no natural lakes with fish > 0.5 km²)
DE POLY (polymictic) STRAT (stratified) DEEP (stratified, deep)
Functional
Functional, < 30 m max. depth
> 30 m max. depth DK L-CB1 (deep)
L-CB2 (shallow) And 7 others (25%)
Stratified, 3-15 m mean depth, alkaline Polymictic, < 3 m mean depth, alkaline and others
EE Not stratified Stratified Soft and dark Soft and bright And 4 others
Functional, avg. hardness
FR Not relevant because of site specific approach
LT Shallow (LCB-2) Interm. Depth (LCB-1) Deep, stratified (LCB-1)
< 3 m mean depth 3-9 m mean depth
> 9 m mean depth NL Shallow, buffered
Deep, buffered Large, deep, buffered Shallow, calcareous Shallow, peat lake
< 3 m mean depth, 0.5 - 100 km², mineral
> 3 m mean depth, 0.5 - 100 km², mineral
> 3 m mean depth, > 100 km², mineral
< 3 m mean depth, 0.5 - 100 km², calcareous
< 3 m mean depth, 0.5 - 100 km², organic PL Polymictic
Stratified
Functional Functional
LV, SK No typology
A2.4 Metrics
Table A 4: Metrics of LFI in the CB GIG and their assignment to the normative criteria of the WFD: spn: species’ number; %N: percentage of total number; NPUE: number per unit of effort; W: weight; %W: percentage of total weight; WPUE: weight per unit of effort.
MS Species composition Abundance Sensitive species Age structure
BE % N specialized spawners % N invertivorous % N omnivorous spn piscivorous %W benthivorous
Tolerance value of species
CZ %W Bream %W Perch %N Ruffe %W Rudd %W Salmonidae
WPUE (> 0+) NPUE (> 0+)
See species comp. Presence 0+ of 6 species
DE %W or %N Bream %W White Bream %W or %N Ruffe %W Pikeperch
WPUE (total) Spn obligatory species Median ind. W of bream, perch, roach
9
MS Species composition Abundance Sensitive species Age structure
%W Perch
%W benthic net species %W benthivorous
Reprod. of stocked species (testing) DK %W Bream + Roach
%W piscivorous
WPUE (per net) See species comp. Average
individual biomass
EE %N Perch logNPUE/ %
nonpisciv.*shoredev)
Simpson DW/log area lake
% filled net sections
FR %N Omnivorous NPUE
WPUE LT %W White Bream
%N Perch %W Benthivorous %W perch+stenoterm.
%W nonnative + translocated
Spn obligatory species
Avg. ind. W of roach
NL %W Bream %W Perch+Roach/
eurytopic
%W phytophilic species %W low oxygen tolerant
← %W of pikeperch
> 40 cm
PL -
L F I + %W Pikeperch %W Pike %W Tench %W Crucian carp %W Perch
% W large Roach in total Roach %W large Bream
%W small Bream
%W large Bream in total Bream %W large Roach
% W White Bream
Total commercial catches
Average for the last 10 years
See species comp.
PL -
C E N %W Pikeperch %W Perch %W Bream %W White Bream %W Roach %W Rudd %W Ruffe %W Tench %W Bleak
BPUE See species comp.
10
A3 WFD compliance
The WFD compliance check was a requirement of the milestone reports in the IC Phase II and is also mentioned in the guidance document (CIS 2011). The compliance criteria should assure that the national systems are in line with the normative definitions of the WFD. The criteria of the milestone reports and the IC guidance are maintained without changes and summarized in Table A 5.
Table A 5: WFD compliance check for the Central Baltic fish systems.
Compliance criteria Compliance checking conclusions
1. Ecological status is classified by one of five classes (high, good, moderate, poor, bad).
Yes for all systems 2. High, good and moderate ecological status are set in
line with the WFD’s normative definitions (boundary setting procedure)
Yes for all systems (although definitions like slight changes, moderate differences or signs of disturbance are unclear)
3. All relevant parameters indicative of the biological quality element are covered. If parameters are missing, Member States need to demonstrate that the method is sufficiently indicative of the status of the QE as a whole.
species composition: Yes for all systems abundance: Yes for all systems
sensitive species: unclear (see Annex AII) age: direct - No, indirect - yes for most systems (see Annex AII)
4. A combination rule of parameters into assessment BQE is defined.
Yes for all systems 5. Assessment is adapted to intercalibration common
types that are defined in line with the typological requirements of the WFD Annex AII and approved by WG ECOSTAT
Yes: DK, LT, NL, no for others
6. The water body is assessed against type-specific near-natural reference conditions
Yes for most systems, FR uses site specific modelling.
7. Assessment results are expressed as EQRs Yes for all systems.
8. Sampling procedure allows for representative information about water body quality/ ecological status in space and time
EN and trawl fishing information is
representative in space (i.e. the whole lake).
Temporal representativeness is under discussion/ investigation. Yes for time and space for Polish LFI+.
9. All data relevant for assessing the biological parameters specified in the WFD’s normative definitions are covered by the sampling procedure
Taxonomic composition: Yes Abundance: Yes
Sensitive species: No for EN only, Yes for multiple gear (NL, PL)
Age: indirectly, 10. Selected taxonomic level achieves adequate
confidence and precision in classification
Yes for all systems.
Comments to the WFD compliance check:
Nr 2 normative definitions of the class boundaries: the normative definitions of WFD are
pretty vague, of course. The Central Baltic LFI were developed to fulfil the demands of the
WFD, and we should suppose that the class boundaries were set with orientation at the
normative definitions. It will be a matter of the IC process to control for similar class boundary
setting among the CB MS.
11
Nr 3 normative definitions of indicative BQE parameters: The Annex V, Nr. 1.2.2 of the WFD provides the normative definitions for high, good and moderate ecological status in lakes. For fish, high status is achieved if (unchanged citation):
1. Species composition and abundance correspond totally or nearly totally to undisturbed conditions.
2. All the type-specific sensitive species are present.
3. The age structures of the fish communities show little sign of anthropogenic disturbance and are not indicative of a failure in the reproduction or development of a particular species.
The normative definition comprises four main aspects of the fish community, i.e. species composition (relative values), species abundance (absolute values), type-specific sensitive species and age structure. As the Central Baltic LFI were developed to fulfil the demands of the WFD, a general compliance should be expected for all of them. However, critical aspects are can be found in the parameters ‘sensitive species’ and ‘age’. The critical aspects are discussed in Annex II. No MS uses a ‘real’ age metric in the national LFI system. Metrics for age (or size or reproduction) are substituted by length or weight parameters. Scientific justification was provided by the CB GIG lead (see Annex II). On this basis, ECOSTAT decided to include national methods in the current intercalibration exercise even if they do not contain age structure metrics (
VAN DEB
UNDet al. 2011).
Nr 5 intercalibration common types: The national typologies are compliant with type-setting criteria mentioned in Annex II of the WFD. Some MS use the IC typology approved by ECOSTAT. For the intercalibration, the CB Lake Fish GIG decided to use another common typology. We do not accept that practical decisions from Phase I should be a decisive prerequisite for a WFD compliance (or successful intercalibration).
Nr 8 representativeness in space and time: The representativeness of different fish sampling gear in time and space is under heavy discussion. First results show that the national systems are able to account for the data variability and provide stable assessment results. We can affirm this point if we change the request for representative sampling to a request for representative assessment results. However, this point will be investigated in detail.
We conclude that all fish assessment systems are WFD compliant.
12
A4 Intercalibration feasibility
Similar to the compliance check, the milestone reports and the guidance for IC Phase II asked for an intercalibration feasibility check and provided a list of questions.
A4.1 Typology as a restricting factor
According to the WFD, the assessment systems have to be based on some kind of typology.
Type descriptors for lakes are given in the directive itself (Annex II): altitude (< 200, 200-800,
> 800 m), mean depth (< 3, 3-15, > 15 m), size (0.5-1, 1-10, 10-100, > 100 km²), and geology (calcareous, siliceous, organic). Based on these criteria an official common typology was set for the Central-Baltic GIG (2008/915/EC 2008; P
OIKANE2009). All types in the CB-GIG are lowland lakes < 200 m altitude. Hydrological water residence time was added to the typology.
Three types were defined for the CB GIG:
L-CB1: shallow, calcareous (3-15 m, > 1 meq/l, 1-10 years of residence time) L-CB2: very shallow, calcareous (< 3 m, > 1 meq/l, 0,1-1 years of residence time) L-CB3: shallow, small, siliceous (3-15 m, 0,2-1 meq/l, 1-10 years of residence time) The CB MS have based their national typologies mainly on the descriptors mean depth and alkalinity. However, other descriptors are added in some cases (max depth, stratification yes/no). Some MS have a fish-specific typology, i.e. the typology is chosen in order to maximize statistical differences of corresponding fish community descriptors (DE)
In order to develop a fish specific typology, Germany has done scientific statistical analyses.
The analyses have shown that the functional criterion ‘stratification’ has the hugest impact on the fish community. A typology based on three types was developed (R
ITTERBUSCH2010;
R
ITTERBUSCHet al. 2010):
POLY: polymictic lakes STRAT: stratified lakes
DEEP: stratified lakes deeper than 30 m max depth
Estonia and Lithuania obtained similar results in their investigations. Unfortunately, the 3 m
/ 15 m mean depth threshold values of the ‘official’ common intercalibration types do not
separate stratified lakes from non-stratified lakes. Therefore, the CB MS decided to use a
typology with a stratification criterion. The typology suggested by Germany was adopted as
fish-specific common intercalibration typology. Poland has also applied a lake typology with
stratification criterion, dividing lakes in polymictic and stratified ones.
13
A4.2 Pressure criteria as restricting factors
All MS have developed systems that take the effect of eutrophication into account, which is the major human pressure in the CB GIG. In many systems additional pressures are evaluated, e.g. human use, shoreline degradation, water level regulation, connectivity …
Some human impacts do affect the fish community, but their relevance in the context of a WFD compliant assessment is still under discussion in CB the Fish group: alien species, translocated species, climate warming. Acidification is not considered relevant on the GIG level, despite some exceptions (e.g. mining lakes). It has to be taken into account, that pressures are highly intercorrelated and often have comparable effects. Intense use will lead to shoreline degradation and eutrophication which will destroy littoral habitat complexity (e.g. in urban areas). Water level regulation also will destroy habitat complexity. As has been mentioned elsewhere, fish are good indicators for ecological status, but comparably bad proxies for single pressures.
It is concluded that the fish assessment is reflecting total pressure intensity and therefore all fish systems are comparable in respect of their pressure indication.
Table A 6: Pressures addressed by the national fish-based lake assessment systems
MS Eutrophication Water level regulation
Shoreline degradation
Combined Comment
BE-F X biotic integrity
(habitat quality, water quality)
CZ X X X X
DE X X X
DK X
EE X overfishing
FR X
NL X X X X
PL X X
LT X X X
UK
A4.3 Sampling comparability
Many MS follow European standard for multimesh-gillnet fishing (EN 14757 2005): CZ, DE, DK, EE, FR, PL follow the EN 14757 more or less exactly, LT excluded small mesh sizes. Although differences might occur due to the deviating application of the EN procedure, the data is generally comparable.
BE-F, NL and PL-LFI use other methods (like trawl, fishery statistics). Data based on different gear is absolutely incomparable to EN data because:
a) Methods sample different habitats,
b) Active (trawl) and passive gears (nets) have different species-specific effectiveness,
c) Selectivity of gears to species or size-classes cannot be converted to other gear,
d) Most systems work with percentages, deviation in one size/species impacts others,
e) The evidence of some species is restricted to certain methods (e.g. littoral species).
14
The fishing methods are not comparable throughout the GIG. This does not imply that the assessment results are not ‘intercalibratable’. We do not want to compare metric values. We want to compare our assessment on a level of the final EQR values and their assignment to status classes. On the other hand, intercalibration methods based on common fish data cannot be applied. Option 1 (common system) and Option 3 (same data acquisition, different systems) have to be ruled out.
A4.4 Metric comparability
Species composition is included in all systems. Abundance is included in all methods, in many cases directly by standardized catches WPUE or NPUE. In some cases, relative abundances of species or functional groups reflect both species composition and abundance traits of the fish communities. Sensitive species are included in most systems, either directly or indirectly. Age is not included or included indirectly. Fish systems generally assess the fish community as a whole, as an integrating BQE for time and place. The total assessment scores are comparable, but the individual metrics are not. Please see comments in Annex AII.
A4.5 Definition of reference conditions and class boundary setting
There are three ways of setting reference conditions and class boundaries (P
OIKANEet al.
unpublished):
1. References: The reference is based on near-natural reference sites. No lakes in true abiotic reference conditions exist in the CB GIG. Therefore, some MS applied the concept of least disturbed conditions (LDC), i.e. the best available lakes in terms of pressure intensities. A second possibility to derive reference conditions is the use of historical data.
Class boundaries are set in comparison to reference conditions.
2. Alternative: Class boundaries are based on sites at similar impairment level
3. Continuous: References and class boundaries are based on pressure-response gradients Table A 7: Benchmarking category and concept of reference condition applied in the
national lake fish indices (LFI). LDC: least disturbed conditions.
MS Benchmark, derivation of reference conditions
BE-F Reference: historical data, lakes of the pike-tench-roach type represent the reference condition CZ Alternative: LDC sites, literature review and expert judgment
DE Continuous/alternative: LDC sites and expert knowledge, pressure-response and site class distribution
DK Reference: paleolimnological data (for trophic reference), equals LDC combined with expert judgment
EE Reference: historical data, expert knowledge, LDC sites FR Continuous: site specific hindcasting method
LT Reference: compared to LDC
NL Reference: LDC sites and expert knowledge
PL Reference: historical data and expert knowledge (LFI+), LDC and expert knowledge (LFI-EN)
Boundaries were set at national level. The boundary setting procedures are heterogeneous and differ between MS, lakes, or even metrics of individual systems (Table A 8). Some possibilities are:
1. regression lines (NL),
2. definition of H/G boundary and consequent equidistant division (NL, FR)
3. value distributions, discontinuities (DE, CZ)
15
In most cases, national expert judgment is included in the class boundary setting. There are no common agreements on abiotic parameters and threshold values representing H/G or G/M boundaries.
Table A 8: National method of class boundary setting. Metric: assignment of scores to metrics, EQR: combination rule for metric scores to obtain a total EQR, class:
assigning ecological status classes to EQR values.
MS Value Setting of class boundaries BE-F Metric
EQR Class
complex, see p. 21
sum of metric scores is transformed to EQR equidistant division of the EQR
CZ Metric
EQR Class
discontinuities of metric values
sum of metric scores is transformed to EQR equidistant division of the EQR
DE Metric
EQR Class
discontinuities of metric values
sum of metric scores is transformed to EQR
least sum of squares between status class and combined pressure index/expert judgment
DK Metric
EQR Class
Based on predefined impact classes sum of metric scores is transformed to EQR Expert judgment
EE Metric
EQR Class
EQR (reference is hindcasted) mean of metric EQR
H/G by expert judgment, others equidistant
FR Metric
EQR Class
EQR (reference is hindcasted) mean of metric EQR
H/G by expert judgment, others equidistant
LT Metric
EQR Class
EQR (reference is 75 % percentile in LDC lakes - see WISER) mean of metric EQR
Discontinuities in pressure-response-relationship, calibrated with expert judgment
NL Metric
EQR Class
EQR (reference is based on LDC sites and expert judgment) weighted sum of metric EQR
Expert judgment based on shifts in fish communities
G/M: loss of habitat for phytophilic fish - change from dominance of phytophilic to dominance of eurytopic;
MP: shift from macrophyte to phytoplankton dominated - shift from dominance of perch/roach to dominance of bream
PL Metric
EQR Class
EQR (reference is based on historical data for LFI+ or LDC and expert knowledge for LFI- EN)
EQR calculated from formulas
Expert judgment based on WFD definitions and shifts in fish communities
16
A5 Summary and conclusions
All CB methods demonstrate that the fish community indication of ecological status summarizes spatial and temporal effects. All systems assess the ecology of the lake as a whole system (including littoral, benthic and pelagic fish). Therefore, habitat specific sampling or the selections of different community characteristics do not represent a major problem. All systems deal with the fact that due to size and complexity of the assessed water body, the ecological status will always be affected by multiple human pressures. Main pressures in CB lakes are eutrophication, human use, water level fluctuation… The pressures may be interdependent and self-enhancing; all of them affect the fish community to some extent. All systems are based on the comparison of the current status with a reference condition although benchmarking procedures differ). The intercalibration seems feasible in terms of assessment concepts.
The main challenges are:
1. Application of different fishing methods which interdicts the application of IC Option 1 (common system) and Option 3 (same data acquisition, different systems)
2. Weak correlation between fish metrics and single pressure parameters.
At the 2
ndIC meeting for Central Baltic Lake Fish Systems, we agreed that intercalibration
should be feasible. However, the options 1 and 3 had to be dropped. The option 2 (different
data acquisition, different numerical evaluation) was chosen. This option requires a common
metric which clearly reacts to the pressures intercalibrated. Unfortunately, the reaction of
fish community metrics to individual pressures usually is low. Furthermore, the development
of a common metric/common system in phase II has turned out unsuitable for the huge
geographical range of the CB GIG. Therefore we decided to make some major modification in
comparison to the description in the phase II guidance: we will not intercalibrate against
individual pressures, but against a total index based on all known pressures. Therefore, it was
agreed to use this combined pressure index as common metric.
17
References
2000/60/EC (2000): Directive 2000/60/EC of the European Parliament and the Council of 23 October 2000 establishing a framework for Community action in the field of water policy vom 22.12.2000. Official Journal of the European Communities Vol 43, L 327 Seiten 1-73 ff.
2008/915/EC (2008): Commission decision establishing, pursuant to Directive 2000/60/EC of the European Parliament and of the Council, the values of the Member State monitoring system classifications as a result of the intercalibration exercise vom 30.10.2008. L 332 Seiten 20-44 ff.
A
PPELBERG, M., B. C. B
ERGQUIST& E. D
EGERMAN(2000): Using fish to assess environmental disturbance of Swedish lakes and streams-a preliminary approach. Verhandlungen der Internationalen Vereinigung für Limnologie 27: 311-315.
A
RGILLIER, C., S. C
AUSSÉ, M. G
EVREY, S. P
ÉDRON, J. B
ORTOLI, S. B
RUCET, M. E
MMRICH, E. J
EPPESEN, T.
L
AURIDSEN, T. M
EHNER, M. O
LIN, M. R
ASK, P. V
OLTA, I. J. W
INFIELD, F. K
ELLY, T. K
RAUSE, A. P
ALM& K. H
OLMGREN(2013): Development of a fish-based index to assess the eutrophication status of European lakes. Hydrobiologia 704: 193-211.
B
ELPAIRE, C., R. S
MOLDERS, I. V. A
UWEELE, D. E
RCKEN, J. B
REINE, G.
VANT
HUYNE& F. O
LLEVIER(2000): An index of biotic integrity characterizing fish populations and the ecological quality of Flandrian water bodies. Hydrobiologia 434: 17-33.
B
IRK, S., W. B
ONNE, A. B
ORJA, S. B
RUCET, A. C
OURRAT, S. P
OIKANE, A. S
OLIMINI, W.
VAN DEB
UND, N.
Z
AMPOUKAS& D. H
ERING(2013): Three hundred ways to assess Europe's surface waters:
An almost complete overview of biological methods to implement the Water Framework Directive. Ecological Indicators 18: 31-41.
B
LABOLIL, P., M. L
OGEZ, D. R
ICARD, M. P
RCHALOVÁ, M. Ř
ÍHA, A. S
AGOUIS, J. P
ETERKA, J. K
UBEČKA& C.
A
RGILLIER(2015): An assessment of the ecological potential of Central and Western European reservoirs based on fish communities. Fisheries Research.
B
OROVEC, J., J. H
EJZLAR, P. Z
NACHOR, J. N
EDOMA, M. Č
TVRTLÍKOVÁ, P. B
LABOLIL, M. Ř
ÍHA, J. K
UBEČKA, D.
R
ICARD& J. M
ATĚNA(2014): Methodology of assessment of the ecological potential heavily modified water bodies and artificial water bodies – lake category (Metodika pro hodnocení ekologického potenciálu silně ovlivněných a umělých vodních útvarů – kategorie jezero). Biology Centre of the AS CR, v.v.i. Institute of Hydrobiology, České Budějovice.
B
REINE, J., G.
V. T
HUYNE& L.
D. B
RUYN(2015): Development of a fish-based index combining data from different type of gear. A case study of reservoirs in Flanders (Belgium). Belgian Journal of Zoology 145: 17-39.
B
REINE, J., P. Q
UATAERT, M. S
TEVENS, F. O
LLEVIER, F. A. M. V
OLCKAERT, E. V
AN DENB
ERGH& J. M
AES(2010): A zone-specific fish-based biotic index as a management tool for the Zeeschelde estuary (Belgium). Marine Pollution Bulletin 60: 1099-1112.
B
RUCET, S., S. P
ÉDRON, T. M
EHNER, T. L. L
AURIDSEN, C. A
RGILLIER, I. J. W
INFIELD, P. V
OLTA, M. E
MMRICH, T. H
ESTHAGEN, K. H
OLMGREN, L. B
ENEJAM, F. K
ELLY, T. K
RAUSE, A. P
ALM, M. R
ASK& E. J
EPPESEN(2013): Fish diversity in European lakes: geographical factors dominate over anthropogenic pressures. Freshwater Biology 58: 1779-1793.
C
AROL, J., L. B
ENEJAM, C. A
LCARAZ, A. V
ILA-G
ISPERT, L. Z
AMORA, E. N
AVARRO, J. A
RMENGOL& E. G
ARCIA- B
ERTHOU(2006): The effects of limnological features on fish assemblages of 14 Spanish reservoirs. Ecology of Freshwater Fish 15: 66-77.
C
ARPENTER, S. R., J. F. K
ITCHELL& J. R. H
ODGSON(1985): Cascading Trophic Interactions and Lake
Productivity. BioScience 35: 634-639.
18
C
AUSSÉ, S., M. G
EVREY, S. P
EDRON, S. B
RUCET, K. H
OLMGREN, M. E
MMRICH, J.
DEB
ORTOLI& C. A
RGILLIER(2012): Fish indicators for ecological status assessment of lakes affected by eutrophication and hydromorphological pressures. WISER deliverable 3.4-4, CEMAGREF, Aix-en-Provence.
CIS (2011): Guidance document on the Intercalibration Process 2008-2011. ECOSTAT 14, Implementation Strategy for the Water Framework Directive (2000/60/EC) - Guidance Document.
D
IEKMANN, M., U. B
RÄMICK, R. L
EMCKE& T. M
EHNER(2005): Habitat-specific fishing revealed distinct indicator species in German lowland lake fish communities. Journal of Applied Ecology 42: 901-909.
E
MMRICH, M., S. B
RUCET, D. R
ITTERBUSCH& T. M
EHNER(2011): Size spectra of lake fish assemblages: responses along gradients of general environmental factors and intensity of lake-use. Freshwater Biology 56: 2316-2333.
E
MMRICH, M., S. B
RUCET, I. J. W
INFIELD, E. J
EPPESEN, T. L
AURIDSEN, P. V
OLTA, C. A
RGILLIER, S. P
EDRON, K.
H
OLMGREN, T. H
ESTHAGEN, D. R
ITTERBUSCH& T. M
EHNER(2012): Changes in size structure of fish assemblages in European lakes along eutrophication and hydromorphological pressure gradients. WISER deliverable 3.4-2.
EN 14757 (2005): European Standard: Water quality - Sampling of fish with multi-mesh gillnets vom 27.06.05. ICS 13.060.70; 65.150.
G
ARCIA, X.-F., M. D
IEKMANN, U. B
RAMICK, R. L
EMCKE& T. M
EHNER(2006): Correlations between type indicator fish species and lake productivity in German lowland lakes. Journal of Fish Biology 68: 1144-1157.
G
ASSNER, H., G. T
ISCHLER& J. W
ANZENBÖCK(2003): Ecological integrity assessment of lakes using fish communities - Suggestions of new metrics developed in two austrian prealpine lakes. International Review of Hydrobiology 88: 635-652.
G
ASSNER, H., J. W
ANZENBÖCK, D. Z
ICK, G. T
ISCHLER& B. P
AMMINGER-L
AHNSTEINER(2005):
Development of a fish based lake typology for natural Austrian lakes >50 ha based on the reconstructed historical fish communities. International Review of Hydrobiology 90: 422-432.
H
EJZLAR, J., J. M
ATĚNA, J. K
OMÁRKOVÁ& J. K
UBEČKA(2003): Typology and reference status standing water bodies: The introductory study for implementation of Water Framework Directive in the Czech Republic (in Czech). Biology Centre of the AS CR, v.v.i. Institute of Hydrobiology.
H
ELMINEN, H., J. K
ARJALAINEN, M. K
URKILAHTI, M. R
ASK& J. S
ARVALA(2000): Eutrophication and fish biodiversity in Finnish lakes. Verhandlungen der Internationalen Vereinigung für Limnologie 27: 194-199.
J
AARSMA, N. (2007): Description of references and metrics for fish in lakes in the Netherlands (unpublished). Witteveen & Bos consulting engineers.
J
EPPESEN, E., J. P. J
ENSEN, M. S
ØNDERGAARD, T. L
AURIDSEN& F. L
ANDKILDEHUS(2000): Trophic structure, species richness, and biodiversity in Danish lakes: changes along phosphorus gradient. Freshwater Biology 45: 201-218.
K
ARR, J. R. (1981): Assessment of Biotic Integrity Using Fish Communities. Fisheries 6: 21-27.
K
ARR, J. R., K. D. F
AUSCH, P. L. A
NGERMEIER, P. R. Y
ANT& I. J. S
CHLOSSER(1986): Assessing biological integrity in running waters - A method and its rationale. Illinois Natural History Survey, Champaign.
K
ELLY, F. L., A. J. H
ARRISON, M. A
LLEN, L. C
ONNOR& R. R
OSELL(2012): Development and application
of an ecological classification tool for fish in lakes in Ireland. Ecological Indicators 18:
19
608-619.
K
UBEČKA, J. (1993): Succession of fish communities in reservoirs of Central and East Europe. In:
M. Straškraba, J. G. Tundisi & A. Duncan (Eds.), Comparative Reservoir Limnology and Water Quality Management, pp. 291. Kluwer Academic Publishers.
L
AUNOIS, L., J. V
ESLOT, P. I
RZ& C. A
RGILLIER(2011a): Selecting fish-based metrics responding to human pressures in French natural lakes and reservoirs: towards the development of a fish-based index (FBI) for French lakes. Ecology of Freshwater Fish 20: 120-132.
L
AUNOIS, L., J. V
ESLOT, P. I
RZ& C. A
RGILLIER(2011b): Development of a fish-based index (FBI) of biotic integrity for French lakes using the hindcasting approach. Ecological Indicators 11: 1572-1583.
L
INDEMANN, R. L. (1942): The trophic-dynamic aspect of ecology. Ecology 23: 399-417.
M
EHNER, T., M. D
IEKMANN, U. B
RÄMICK& R. L
EMCKE(2005): Composition of fish communities in German lakes as related to lake morphology, trophic state, shore structure and human-use intensity. Freshwater Biology 50: 70-85.
M
EHNER, T., K. H
OLMGREN, T. L. L
AURIDSEN, E. J
EPPESEN& M. D
IEKMANN(2007): Lake depth and geographical position modify lake fish assemblages of the European 'Central Plains' ecoregion. Freshwater Biology 52: 2285-2297.
P
ALM, A., T. K
RAUSE, A. J
ÄRVALT& M. S
ILM(2012): Cyprinids in Estonian small lakes: comparison between main water bodies. Balwois Proceedings 5: 1-9.
P
OIKANE, S. (2009): Water Framework Directive intercalibration technical report - Part 2: Lakes.
Joint Research Center.
P
OIKANE, S., S. B
IRK, J. B
ÖHMER, H. G
ASSNER, S. H
ELLSTEN, C.
D. H
OYOS, M. K
ELLY, A. L. S
OLHEIM, M. O
LIN, K. P
ALL, G. P
HILLIPS, R. P
ORTIELJE, D. R
ITTERBUSCH, L. S
ANDIN, A. K. S
CHARTAU, A. S
OLIMINI, M.
V.
D
. B
ERG, G. W
OLFRAM& W.
V.
D. B
UND(unpublished): Lake ecological assessment:
comparing the incomparable?
P
RCHALOVÁ, M., J. K
UBEČKA, M. Č
ECH, J. F
ROUZOVÁ, V. D
RAŠTÍK, E. H
OHAUSOVÁ, T. J
ŮZA, M. K
RATOCHVÍL, J. M
ATĚNA, J. P
ETERKA, M. Ř
ÍHA, M. T
UŠER& M. V
AŠEK(2009): The effect of depth, distance from dam and habitat on spatial distribution of fish in an artificial reservoir. Ecology of Freshwater Fish 18: 247-260.
P
RCHALOVÁ, M., J. K
UBEČKA, M. V
AŠEK, J. P
ETERKA, J. S
ED’
A, T. J
ŮZA, M. Ř
ÍHA, O. J
AROLÍM, M. T
UŠER, M.
K
RATOCHVÍL, M. Č
ECH, V. D
RAŠTÍK, J. F
ROUZOVÁ& E. H
OHAUSOVÁ(2008): Distribution patterns of fishes in a canyon-shaped reservoir. Journal of Fish Biology 73: 54-78.
R
ITTERBUSCH, D. (2010): First LakeFish Intercalibration Meeting30.11.-01.12.2010, Minutes, Berlin: 9.
R
ITTERBUSCH, D. (2011): Proposal of an alternative option to assure that national good/moderate class boundaries correspond to comparable levels of ecosystem alteration. Draft discussed at the CB LakeFish meeting on November 18, 2011 Institut für Binnenfischerei e. V., Potsdam-Sacrow.
R
ITTERBUSCH, D., U. B
RÄMICK& T. M
EHNER(2010): Development of a German lake typology for fish-based assessments of ecological status/integrity. Institut für Binnenfischerei e.V., Potsdam-Sacrow.
R
ITTERBUSCH, D., U. B
RÄMICK& T. M
EHNER(2014): A typology for fish-based assessment of the ecological status of lowland lakes with description of the reference fish communities.
Limnologica - Ecology and Management of Inland Waters 49: 18-25.
Š
MEJKAL, M., D. R
ICARD, M. P
RCHALOVÁ, M. Ř
ÍHA, M. M
UŠKA, P. B
LABOLIL, M. Č
ECH, M. V
AŠEK, T. J
ŮZA,
A. M. H
ERRERAS, L. E
NCINAD, J. P
ETERKA& J. K
UBEČKA(2015): Biomass and abundance biases
in European standard gillnet sampling. PLoS ONE 10.
20
S
ØNDERGAARD, M., E. J
EPPESEN, J. P. J
ENSEN& S. L. A
MSINCK(2005): Water Framework Directive:
ecological classification of Danish lakes. Journal of Applied Ecology 42: 616-629.
S
ØNDERGAARD, M., T. L
AURIDSEN, E. K
RISTENSEN, A. B
ATTRUP-P
EDERSEN, P. W
IBERG-L
ARSEN, R. B
JERRING& N. F
RIBERG(2013): Biologiske indikatorer til vurdering af økologisk kvalitet i danske søer og vandløb. 59, Aarhus Universitet, Aarhus.
VAN DE
B
UND, W., W. B
ONNE, S. P
OIKANE& N. Z
AMPOUKAS(2011): Method compliance checking:
which methods can be included in the intercalibration exercise? Document discussed at the ECOSTAT meeting of 30.03.2011 V 2.0, Joint Research Center.
V
RBA, J., J. K
OPÁČEK& J. F
OTT(2000): Long-term limnological research of the Bohemian Forest lakes and their recent status. Silva Gabreta 4: 7-28.
W
HITTIER, T. R. (1999): Development of IBI metrics for lakes in southern New England. In: T. P.
Simon (Ed.) Assessing the sustainability and biological integrity of water resources
using fish communities, pp. 563-582. CRC-Press.
21
Annex AI: National descriptions of Lake fish assessment systems Belgium-Flanders
All lakes used for the index development are artificial or heavily modified and can be considered as polymictic. Only some lakes are connected to a river. The fish protocol used in Flanders in lakes is one overnight fyke/ha for two successive days with a minimum of 4 and a maximum of 20 per lake combined with electric fishing along 250m-long shore transects per ha.
Index development
Fish were attributed to guilds based on literature review. Species were categorized according to their tolerance for oxygen deficiency and habitat structure degradation. Tolerance scores from 1 (tolerant) to 5 (intolerant) were given to each species. For each lake, gear specific metric values were calculated using reference species only.
Statistical analyses
We assessed the number of species and candidate metrics. To retrieve less-skewed distributions percentage metrics were square-root transformed and count metrics were log- transformed (logx+1). Diversity metrics were kept untransformed.
First correlation among pressure scores was assessed (measure of association, p (Fisher)) to avoid colinearity. Pearson correlation was applied to assess correlation between lake depth and lake surface (log x+1) transformed values.
The response of metrics to pressures (log transformed values) was analysed with linear mixed regression models. As some locations were sampled several times we added locality and year as random effects. We started with a full model including all pressures and applied a stepwise backward selection until only significant terms remained. Normality assumptions were assessed with residual plots. Redundancy of responsive metrics was analysed with a Pearson correlation. Model fit and expert judgment, when needed, was used to select one metric among the redundant ones. The statistical package used was R.2.15.2 (R Development Core Team, 2012).
Threshold value determination for the selected metrics followed Breine et al. (B
REINEet al.
2010). First, the GEP class boundary was defined:
For metrics assessing number of species 80% of the reference number was taken as the GEP status threshold value.
For the relative percentage metrics (Mpi metrics) we calculated the ratio of the species occurrence included in a particular metric over the total number of species in the reference list. This value is used as the GEP threshold.
Once the GEP is defined the other integrity classes are defined by applying trisection with GEP values.
The average value from the highest impacted sites (total pressure >7) was used to define the minimum percentage weight of benthivorous species (BenWei) and the bream and roach associated metric (AbrRut).
The sum of the metric scores obtained with each method gives the index of biotic integrity
(IBI) score for a particular lake. To comply with the WFD this score is transformed to an
22
ecological quality ratio (EQR) calculated as a value between 0 and 1: EQR = (IBI -lowest IBI possible)/(maximum IBI possible-lowest IBI possible). The EQR for the MEP status is 1 under which four classes are defined: GEP (lower threshold value 0.75), moderate (0.5), poor (0.25) and bad (<0.25). The transformation to equal interval classes is obtained using following formula for each integrity interval (piecewise transformation):
T EQR = LV T EQR + (O EQR - LV O EQR)/(UV O EQR - LV O EQR)*0.25
O and T stand for original and transformed EQR value, UV and LV (upper and lower value of integrity class). When during one campaign more than one site is assessed within one lake, data obtained with same method is summed and transformed to catch per unit effort (i.e. per m² or per fyke day) to calculate the final EQR for the lake. Allowing a class difference of one unit indices were validated by comparing the integrity class obtained per lake with its assessed pressure status. The pressure status appreciation is obtained by applying threshold to the pressure scores: bad (7-9), poor (4-6), moderate (3), GEP (2) and MEP (0-1). We assessed data of lakes used for the index development and an independent set of data consisting of fish data from eight lakes not included in the index development.
Metrics and threshold values
MpiSpa: Percentage of specialized spawners: species composition and richness (electric fish data)
MpiInv: Percentage of invertivorous individuals: trophic composition (electric fish data)
MpiOmn: Percentage of omnivorous individuals: trophic composition (fyke net data) MnsPis: Number of piscivorous species: trophic composition (fyke net data)
BenWei: Percentage weight of benthivore: species trophic composition (fyke net data) ManTol: Tolerance value species: composition and richness (fyke net data)
Additional explanations: Mpi values are relative e.g. 100 individuals are caught and 5 of those are omnivorous then MpiOmn=5/100= 5%, this is so for all Mpi metrics. (Please note that the GEP boundary was determined by comparing the % of species to a reference list of species while the metric is calculated as the percentage of individuals in the total catch). ManTol is the sum of the tolerance value (only species on the reference list are taken into account) of all species that are caught within one survey. E.g. 4 species were caught having a tolerance value of 5; 3 and 1; and one species not on the reference list than the ManTol value= 5+3 +1 or 8.
MEP status is obtained when all 21 reference species are present: Abramis brama (Linnaeus,
1758), Anguilla anguilla (Linnaeus, 1758), Blicca bjoerkna (Linnaeus, 1758), Carassius
carassius (Linnaeus, 1758), Carassius gibelio (Bloch, 1782), Cyprinus carpio carpio (Linnaeus,
1758), Esox lucius (Linnaeus, 1758), Gasterosteus aculeatus (Linnaeus, 1758), Gobio gobio
(Linnaeus, 1758), Gymnocephalus cernua (Linnaeus, 1758), Leucaspius delineatus (Heckel,
1843), Leuciscus idus (Linnaeus, 1758), Lota lota (Linnaeus, 1758)**, Perca fluviatilis
(Linnaeus, 1758), Pungitius pungitius (Linnaeus, 1758), Rhodeus sericeus (Pallas, 1776), Rutilus
rutilus (Linnaeus, 1758), Sander lucioperca (Linnaeus, 1758), Scardinius erythrophthalmus
(Linnaeus, 1758), Silurus glanis (Linnaeus, 1758)**, Tinca tinca (Linnaeus, 1758).
23
For the GEP status we should have a relative percentage of specialized spawners between 28.5 and 21.4%. In the moderate status percentages range between 21.4 and 14.2% or when it is higher than 28.5%.The species considered are: pike, gudgeon, burbot, ruffe, rudd and tench (Table A 9).
The relative percentage of invertivorous species (perch (<13 cm total length), ruffe and gudgeon) in GEP ranges between 28.9 and 14.2% and in the moderate status between 14.2 and 9.4% or when it is higher than 28.9%.
The omnivorous species include three-spined stickleback, eel, tench, bream, Prussian carp, common carp, ide, ninespine stickleback, roach and rudd. In GEP their relative percentage ranges between 15.9 and 7.9%. The moderate status is achieved when it is less than 7.9% or ranges between 15.9 and 31.7%.
Five species are considered as piscivorous: burbot, wels catfish, pike-perch, perch (≥ 13cm total length) and pike. GEP is obtained if 3 or 4 of these are present. If only 2 are present the lake is in the moderate status for this metric.
The relative weight percentage of benthivore species should range between 14 and 7% to achieve the GEP status. If less than 7% then we have the moderate status which is also obtained when the percentage ranges between 28 and 14. Species considered here are:
bream, white bream, common carp, ruffe and tench. Finally the GEP status is obtained when
17 species of the reference list are present (listed above).
24
Table A 9: Selected metrics for Flemish lakes and their threshold values for the metric and EQR-scores.
MEP GEP Moderate Poor Bad
Electrofishing data
Metric - score 1 0.8 0.6 0.4 0.2
% Specialised spawners < 28.5 ≥ 21.4 ≥28.5 & < 21.4≥ 14.2 < 14.2 ≥ 7.1 < 7.1
% Invertivorous
individuals < 28.9 ≥ 14.2 ≥ 28.9 & < 14.2 ≥ 9.4 < 9.4 ≥ 4.7 < 4.7 Fyke net data
Metric - score 1 0.8 0.6 0.4 0.2
% Omnivores < 15.9 ≥7.9 < 31.7 ≥ 15.9 & <7.9 < 47.6 ≥ 31.7 ≥ 47.6 Number of piscivorous
species 5 <5 ≥ 3 2 1 0
% Weight of benthivore
species < 14 ≥ 7 < 28.0 ≥ 14.0 & <7 < 42.0 ≥ 28.0 ≥ 42.0
Tolerance value 50 <50 ≥40 <40 ≥27 <27 ≥13 <13
EQR 1 < 1 ≥ 0.75 < 0.75 ≥ 0.50 < 0.50 ≥ 0.25 < 0.25
Appreciation MEP GEP Moderate Poor Bad
Pressure-response
The response of metrics to pressures (i.e. industry, agriculture activity, shore modification and development constructions log transformed values to meet requirements of linear models) and predictors (depth, surface, trees) was analysed with linear mixed regression models. As some locations were sampled several times we added locality and year as random effects. We started with a full model including all pressures and predictors. We applied a stepwise backward selection until only significant terms remained. Normality assumptions were assessed with residual plots. To define the goodness-of-fit, the marginal and conditional R² for each fitted model were calculated as described by NAKAGAWA & SCHIELZETH (2013).
Only the metric response to pressures was decisive for the selection (R² conditional>35%).
Results in Table A 10.
Redundancy of responsive metrics was analysed with a Pearson correlation. To choose among the correlated metrics (c ≥0.7; p ≤0.001), the one with the best fitted model was taken.
Secondly, among the less correlated metrics (c <0.7 and ≥0.5; p ≤0.05), the one that least
correlates with other metrics was selected. The results are shown in Table A11.
25
Table A 10: Reaction of metrics with uncorrelated pressures in reservoirs. (Surlake: reservoir surface; Depth: average depth of reservoir; Dev:
percentage of construction; Agr: percentage of agriculture activities; Tree: percentage of trees: Nat: percentage of natural banks) described metrics (log (L) or square root (SR) transformed.
model <-lmer(metric ~ Lake surface + Development + Depth + Natural banks + Agriculture + Trees + (1|reservoir) + (1|year))
Metrics (E) Selected model
p value variable 1
p value variable 2
p value
variable 3 R² Mar R² Cond
LMnsInv 0.460-0.048Tree 0.0154 0.244 0.528
SRMpiSpa 3.177+0.125Nat-0.612Tree 0.0044 0.0244 0.193 0.363
SRManRec 5.786+0.597Agr 0.0485 0.085 0.136
SRMpiOmn 8.384-0.181Depth 0.0008 0.277 0.404
SRMpiPis 4.576+0.193Depth-1.243Tree+0.979Nat 0.0060 0.0234 0.0472 0.264 0.583
SRMpiInv 4.869-1.272Tree+0.144Depth+1.012Nat 0.0101 0.0135 0.0323 0.209 0.523
SRAbrRut 0.3444-0.183Depth 0.0155 0.212 0.360
SRBenWei 1.196-1.775Agr-0.741Dev 0.0002 0.0181 0.254 0.275
SRSanLuc 0.259-0.101Depth+0.426Nat 0.0370 0.0940 0.083 0.168
SRPerFlu 0.346+0.033Surlake+0.124Depth+0.659Dev 0.0038 0.0041 0.0116 0.274 0.282
LManTol 0.622+0.005Depth 0.0717 0.091 0.276
Metrics (F) Selected model
LMnsTot 0.503+0.18Tree-0.016Depth 0.0007 0.0042 0.358 0.741
LManBio 2.5-0.576Tree-0.031Depth-0.006Surlake 0.0001 0.0040 0.0060 0.165 0.310
LMnsPis 0223+0.056Nat 0.0450 0.139 0.539
SRMpiSpa 1.901+0.187Tree-0.351Surlake-0.401Nat 0.0005 0.0006 0.0020 0.145 0.579
SRMpiOmn 2.021+1.268Agr+1.352Tree 0.0004 0.0024 0.281 0.390
SRMpiPis 3.979-0.316Depth-0.116Surlake 0.0098 0.0341 0.296 0.523
SRMpiInv 6.482-1.591Tree+0.034Surlake 0.0168 0.0495 0.221 0.532
SRAbrRut -0.196+1.322Tree 0.0090 0.257 0.644
SRBenWei -0.647+1.219Agr+1.288Tree 0.0036 0.0184 0.296 0.502
26
model <-lmer(metric ~ Lake surface + Development + Depth + Natural banks + Agriculture + Trees + (1|reservoir) + (1|year))
Metrics (E) Selected model
p value variable 1
p value variable 2
p value
variable 3 R² Mar R² Cond
SRSanLuc -0.453+0.889Tree 0.0310 0.167 0.468
LManTol 0.599-0.044Dev+0.068Tree 0.0150 0.0220 0.268 0.539
Table A 11: Pearson coefficient (c) and significance (**p ≤0.001; * p≤0.05) for correlation analysis of model fitted metrics with electric and fyke data.
Electric MnsInv MpiSpa ManRec MpiOmn MpiPis MpiInv AbrRut BenWei PerFlu SanLuc
MpiSpa 0.0481 1
ManRec 0.0965 0.2788* 1
MpiOmn -0.1699 0.0205* -0.1964* 1
MpiPis 0.2766* 0.2166* 0.1864* -0.7123** 1
MpiInv 0.3700** 0.1153 0.2349* -0.7051** 0.9266** 1
AbrRut 0.0334 -0.2274* 0.2003* 0.2955* -0.1937* -0.0756 1
BenWei 0.1654 0.2111 0.1412 0.0456 -0.1139 -0.0401 0.018 1
PerFlu 0.1182 -0.0756 0.2642* -0.5314** 0.6938** 0.6993** 0.0561 -0.2327* 1
SanLuc 0.1363 -0.0243 -0.1897* 0.0976 -0.0158 -0.0602 -0.0003 -0.0673 -0.2387* 1
Mantol 0.1330 0.4280* -0.0267 -0.0716 0.1904* 0.0941 -0.2247* 0.0290 0.03782 -0.3105*
Fykes MnsTot ManBio MnsPis MpiSpa MpiOmn MpiPis MpiInv AbrRut BenWei SanLuc
ManBio 0.8138** 1
MnsPis 0.5750** 0.4796** 1
MpiSpa 0.2657** 0.1303* -0.0404 1
MpiOmn 0.4891** 0.6088** 0.1635 0.2625** 1
MpiPis -0.2866** -0.2991** 0.5132** -0.3236** -0.5729** 1
27
Fykes MnsTot ManBio MnsPis MpiSpa MpiOmn MpiPis MpiInv AbrRut BenWei SanLuc
MpiInv 0.0878 0.0391 0.3967** -0.1591* -0.2003* 0.5904** 1
AbrRut 0.5928** 0.3390* 0.2711** 0.0900 0.3486** -0.1780* -0.0672 1
BenWei 0.5391** 0.4457** 0.1795 0.1892* 0.4060** -0.3292** -0.1071 0.4883** 1
SanLuc 0.2201** 0.1935* 0.3908** -0.1974* -0.0400 0.2814** -0.2182* 0.1029 0.0575 1
ManTol 0.3506** 0.4505** 0.3795** 0.2322* 0.4002** 0.1821* 0.5469** 0.0405 0.1603 -0.1484*
28
